The combination of a thermogravimetric analyzer (TGA) with mass spectrometry (MS) to analyze evolved gases is a well-known
technique. However, TG–MS does not allow differentiation between gases that evolve simultaneously; TG combined with gas chromatography–mass
spectrometry (TG–GC–MS) can allow for a more complete characterization.
Pyrolysis is now a standard way to introduce samples to gas chromatography (GC) systems, but it is limited because it cannot
correlate the evolution of volatiles to precise temperatures. The same issue arises in thermal analysis when using data from
a thermogravimetric analyzer (TGA). A TGA measures the change in the weight of a sample as a function of temperature. The
temperature-to-weight loss relationship is precisely defined, but researchers are limited to guessing what is eluting at each
temperature point. Replacing a headspace or pyrolysis unit with a TGA can allow a more precise determination of the off-gas's
relationship to temperature. This can help analysts to determine the source of the off-gas, whether it is the result of evaporation
or the combination of another material.
(PHOTO CREDIT: EASTNINE INC./GETTY IMAGES)
The use of a TGA also allows characterization of materials not suitable for gas chromatography–mass spectrometry (GC–MS) including
solid residues left after pyrolysis. For example, fillers such as carbon black, glass fibres, or oxides; or the heavy oils
remaining after a simulated distillation.
Researchers can also produce heat capacity and enthalpy values for a reaction that is evolving gas by using a simultaneous
differential thermal analyzer (STA) to introduce a sample to a gas chromatograph system. This can be extremely useful in combustion
reactions such as the burning of a biofuel or in the release of a chemical during a condensation polymerization reaction.
Using TG to Understand the Leaching Process
GC–MS can differentiate between similar compounds, such as methanol and ethanol, and can also characterize complex gaseous
mixtures coming off a sample. GC–MS can be used to detect extremely small levels of contaminants such as those in leachates.
In fact, information can be collected on the small amounts of material evolved from complex matrices and this ability to analyze
very small amounts of evolved gases is needed to understand processes like leaching, where a packaging material and/or its
contents contaminate each other and degrade properties. This is a concern for many people as bisphenol A (BPA) and other plastizers
can be leached into food. In some cases, researchers can even quantify the size of the effect by determining the amount of
leachable material absorbed by using gas sample capture loops. This is often semi-quantitative as precise determinations can
In the following two case studies, we look at the benefits of combining the techniques.
Case Study: Analysis of Coffee Beans in Plastic Containers
TG combined with GC–MS can be applied to the detection of leachable substances that contaminate foods and pharmaceuticals
and which are often present at very low levels and in complex matrices, such as plasticizers or other packaging contaminants.
It is important to understand the relationship of temperature with the evolution of the gas, to avoid misidentifying the products
of decomposition or pyrolysis of the sample as a contaminant.
Method: The analysis was performed on a Pyris 1 TGA (PerkinElmer) using alumina pans and a standard furnace. The instrument was calibrated
with nickel and iron and all samples were run under helium purge. Heating rates varied from 5 °C/min to 40 °C/min, depending
on the sample under test. The furnace was burned off between runs in air. Samples were approximately 10–15 mg. Data analysis
was performed using Pyris 9.0 Software (PerkinElmer). During the TG–GC–MS analysis, a Clarus 680 C GC–MS (PerkinElmer) was
used. A 0.32 mm i.d. deactivated fused-silica transfer line was connected to the GC injector port. The transfer line was heated
to 210 °C and connected to a 30 m × 0.32 mm, 1-µm Elite WAX column (PerkinElmer). In both cases, data analysis was performed
using TurboMass GC–MS software (PerkinElmer).
Results: Figure 1 shows the thermogram from the analysis of two types of coffee beans. The TGA weight loss curve for both is similar
and uncomplex, but the derivative curve of weight loss versus temperature gives an indication of the complexity of the material.
Figure 1: Thermogram from the analysis of African (blue) and Indonesian (red) green coffee beans.
The derivative curve can be used as a "fingerprint" to identify the type of bean, because there are distinct derivative profiles
for the Ethiopian bean (blue) compared to the Indonesian bean (red). Figure 2 shows the higher retention time peaks, with
a caffeine peak at 60.8 min as well as a phthlate peak at 56 min.
Figure 2: Chromatogram of the off-gas from the TGA run. Caffeine appears at 60.8 min.